<![CDATA[Particle Swarm Optimization (PSO) is a widely used metaheuristic approach for solving optimization problems. Recent developments in this field involve the adaptation of human learning behaviors to enhance algorithmic performance. One such adaptation is the Adaptive Particle Swarm Optimization based on Human Social Learning (APSO-HSL), a variant of PSO that incorporates human inspired learning strategies. This study aims to enhance the performance of APSO-HSL on the Traveling Salesman Problem (TSP) by incorporating additional human learning strategies. The proposed algorithm, named Modified Adaptive Particle Swarm Optimization–Human Learning Strategies (MAPSO-HLS), integrates learning mechanisms from Human Learning Optimization (HLO), including individual, random, and social learning. This research is classified as applied research and algorithmic experimentation, focusing on the development and modification of a metaheuristic algorithm to solve a well-known combinatorial optimization problem. Benchmark datasets from the Traveling Salesman Problem Library (TSPLIB) are used for evaluation, and all computations and experiments are implemented in Python. The performance of MAPSO-HLS is compared with the exact method in terms of shortest distance and computation time. The results of the study indicate that the MAPSO-HLS algorithm is capable of producing TSP solutions with low total distance deviation, below 10%, compared to exact solutions across all tested datasets. This reflects a high level of solution accuracy. In addition, MAPSO-HLS demonstrates better time efficiency than the exact ILP method, particularly for datasets with a large number of cities. The integration of human learning strategies within the adaptive PSO framework provides significant advantages in terms of both efficiency and effectiveness in solving TSP.]]>
